Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
handling Flash 11.3 regressions
207 views
Skip to first unread message
Robert O'Callahan
Jul 22, 2012, 10:16:11 PM7/22/12
to mozilla.dev.planning group
The Flash 11.3 update has caused some very serious problems in Firefox.
Repainting issues (bug 762948), issues resizing the browser window (bug
765788), and various crashes. I've heard reports of a spike in related
activity on SUMO and reports of a lot of users leaving Firefox over these
issues. So far these seem to be Adobe's bugs, probably related to their new
sandbox since that is Firefox only (maybe Opera too?) and at least some of
these bugs go away if the user disables the sandbox. We've looked into
workarounds for some of them but it's not easy. Adobe is responding but not
very quickly.
Is anyone focused on this problem? Have we considered something drastic
like forcibly disabling their sandbox mode until these bugs are fixed?
Rob
--
“You have heard that it was said, ‘Love your neighbor and hate your enemy.’
But I tell you, love your enemies and pray for those who persecute you,
that you may be children of your Father in heaven. ... If you love those
who love you, what reward will you get? Are not even the tax collectors
doing that? And if you greet only your own people, what are you doing more
than others?" [Matthew 5:43-47]
Repainting issues (bug 762948), issues resizing the browser window (bug
765788), and various crashes. I've heard reports of a spike in related
activity on SUMO and reports of a lot of users leaving Firefox over these
issues. So far these seem to be Adobe's bugs, probably related to their new
sandbox since that is Firefox only (maybe Opera too?) and at least some of
these bugs go away if the user disables the sandbox. We've looked into
workarounds for some of them but it's not easy. Adobe is responding but not
very quickly.
Is anyone focused on this problem? Have we considered something drastic
like forcibly disabling their sandbox mode until these bugs are fixed?
Rob
--
“You have heard that it was said, ‘Love your neighbor and hate your enemy.’
But I tell you, love your enemies and pray for those who persecute you,
that you may be children of your Father in heaven. ... If you love those
who love you, what reward will you get? Are not even the tax collectors
doing that? And if you greet only your own people, what are you doing more
than others?" [Matthew 5:43-47]
akeybl
Jul 22, 2012, 11:15:57 PM7/22/12
to rob...@ocallahan.org, dev-pl...@lists.mozilla.org
Robert,
The release management, stability, SUMO, metrics, and QA teams at Mozilla along with engineering (bsmedberg, etc) have all been actively engaged on issues related to, or suspected to be related to, Adobe Flash 11.3 since its release. Comments in the stability bugs you reference should reflect that. Similarly, Adobe has multiple teams engaged on 11.3 issues, and they are analyzing all actionable data we provide, working to fix as many of the top issues in new versions of Flash as possible.
We weigh our options on a near daily basis, and update our support documentation with the best options for our affected users (including disabling protected mode). As you suggest, drastic solutions are drastic and should be avoided until all other better options have been exhausted.
Rest assured that Flash 11.3 issues are getting the proper attention at Mozilla and will continue to until their resolution.
-Alex
Robert O'Callahan <rob...@ocallahan.org> wrote:The Flash 11.3 update has caused some very serious problems in Firefox.
_______________________________________________
dev-planning mailing list
dev-pl...@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-planning
The release management, stability, SUMO, metrics, and QA teams at Mozilla along with engineering (bsmedberg, etc) have all been actively engaged on issues related to, or suspected to be related to, Adobe Flash 11.3 since its release. Comments in the stability bugs you reference should reflect that. Similarly, Adobe has multiple teams engaged on 11.3 issues, and they are analyzing all actionable data we provide, working to fix as many of the top issues in new versions of Flash as possible.
We weigh our options on a near daily basis, and update our support documentation with the best options for our affected users (including disabling protected mode). As you suggest, drastic solutions are drastic and should be avoided until all other better options have been exhausted.
Rest assured that Flash 11.3 issues are getting the proper attention at Mozilla and will continue to until their resolution.
-Alex
Robert O'Callahan <rob...@ocallahan.org> wrote:The Flash 11.3 update has caused some very serious problems in Firefox.
dev-planning mailing list
dev-pl...@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-planning
Robert O'Callahan
Jul 22, 2012, 11:18:08 PM7/22/12
to akeybl, dev-pl...@lists.mozilla.org
Great, thanks for the reassurance :-).
Dao
Jul 23, 2012, 5:20:03 AM7/23/12
to ake...@mozilla.com, rob...@ocallahan.org
On 23.07.2012 05:15, akeybl wrote:
> We weigh our options on a near daily basis, and update our support documentation with the best options for our affected users (including disabling protected mode).
Not all affected users will consult SUMO, will they? Probably not even a
> We weigh our options on a near daily basis, and update our support documentation with the best options for our affected users (including disabling protected mode).
majority. Seems like this will mitigate the problem, never really solve it.
> As you suggest, drastic solutions are drastic and should be avoided until all other better options have been exhausted.
Robert Kaiser
Jul 23, 2012, 9:39:08 AM7/23/12
to
Dao schrieb:
within the Flash plugin itself. And Adobe probably would consider this a
security matter as the sandbox ("Protected Mode") is a security feature.
I guess they would do everything possible not to disable a security feature.
That said, as Alex has laid out, teams on our side and on Adobe's side
are working on improving the situation and are talking to each other
regularly.
We also still need steps to reproduce on a number of the crash an hang
problems, please looks at the corresponding bugs if you have an issue
you can reproduce, both our developers and the people at Adobe are
always happy to get those steps so they can work on fixing things even
better.
I'm personally worried that the problems around Flash 11.3 with
Protected Mode could have cost us users, but we don't have proof of
that, one reason being that there's a drop in user numbers in summer anyhow.
In the end, what I have learned in the recent weeks is that we need to
work even harder on making HTML5 work smooth enough that we hopefully
need Flash less and less in the future, esp. when it comes to games on
the web.
Robert Kaiser
>> As you suggest, drastic solutions are drastic and should be avoided
>> until all other better options have been exhausted.
>
> Really? Would we lose much by disabling the sandbox for now?
From what I understand, it's not really our decision, as that's buried
>> until all other better options have been exhausted.
>
> Really? Would we lose much by disabling the sandbox for now?
within the Flash plugin itself. And Adobe probably would consider this a
security matter as the sandbox ("Protected Mode") is a security feature.
I guess they would do everything possible not to disable a security feature.
That said, as Alex has laid out, teams on our side and on Adobe's side
are working on improving the situation and are talking to each other
regularly.
We also still need steps to reproduce on a number of the crash an hang
problems, please looks at the corresponding bugs if you have an issue
you can reproduce, both our developers and the people at Adobe are
always happy to get those steps so they can work on fixing things even
better.
I'm personally worried that the problems around Flash 11.3 with
Protected Mode could have cost us users, but we don't have proof of
that, one reason being that there's a drop in user numbers in summer anyhow.
In the end, what I have learned in the recent weeks is that we need to
work even harder on making HTML5 work smooth enough that we hopefully
need Flash less and less in the future, esp. when it comes to games on
the web.
Robert Kaiser
Dao
Jul 23, 2012, 10:03:02 AM7/23/12
to
On 23.07.2012 15:39, Robert Kaiser wrote:
> Dao schrieb:
>>> As you suggest, drastic solutions are drastic and should be avoided
>>> until all other better options have been exhausted.
>>
>> Really? Would we lose much by disabling the sandbox for now?
>
> From what I understand, it's not really our decision, as that's buried
> within the Flash plugin itself.
Maybe. I was working with roc's assumption that there would be a way for
> Dao schrieb:
>>> As you suggest, drastic solutions are drastic and should be avoided
>>> until all other better options have been exhausted.
>>
>> Really? Would we lose much by disabling the sandbox for now?
>
> From what I understand, it's not really our decision, as that's buried
> within the Flash plugin itself.
us to disable the sandbox.
> And Adobe probably would consider this a
> security matter as the sandbox ("Protected Mode") is a security feature.
> I guess they would do everything possible not to disable a security
> feature.
Protected Mode.
> That said, as Alex has laid out, teams on our side and on Adobe's side
> are working on improving the situation and are talking to each other
> regularly.
>
> We also still need steps to reproduce on a number of the crash an hang
> problems, please looks at the corresponding bugs if you have an issue
> you can reproduce, both our developers and the people at Adobe are
> always happy to get those steps so they can work on fixing things even
> better.
and a way to address this regression using the sledgehammer, doing just
that may be better than waiting until we found a way to reproduce each
problem.
> I'm personally worried that the problems around Flash 11.3 with
> Protected Mode could have cost us users, but we don't have proof of
> that, one reason being that there's a drop in user numbers in summer
> anyhow.
drop is massive enough for us to see it, it can't be undone anymore. We
should try to prevent the drop.
Martijn
Jul 23, 2012, 12:47:53 PM7/23/12
to Robert Kaiser, dev-pl...@lists.mozilla.org
On Mon, Jul 23, 2012 at 3:39 PM, Robert Kaiser <ka...@kairo.at> wrote:
> We also still need steps to reproduce on a number of the crash an hang
> problems, please looks at the corresponding bugs if you have an issue you
> can reproduce, both our developers and the people at Adobe are always happy
> to get those steps so they can work on fixing things even better.
There is at least
> We also still need steps to reproduce on a number of the crash an hang
> problems, please looks at the corresponding bugs if you have an issue you
> can reproduce, both our developers and the people at Adobe are always happy
> to get those steps so they can work on fixing things even better.
https://bugzilla.mozilla.org/show_bug.cgi?id=775646
with a testcase that makes Flash crash reproducible for me.
I hope someone from the Adobe team can take a look and comment.
Regards,
Martijn
> I'm personally worried that the problems around Flash 11.3 with Protected
> Mode could have cost us users, but we don't have proof of that, one reason
> being that there's a drop in user numbers in summer anyhow.
>
> In the end, what I have learned in the recent weeks is that we need to work
> even harder on making HTML5 work smooth enough that we hopefully need Flash
> less and less in the future, esp. when it comes to games on the web.
>
> Robert Kaiser
>
> _______________________________________________
> dev-planning mailing list
> dev-pl...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-planning
--
> dev-planning mailing list
> dev-pl...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-planning
Martijn Wargers - Help Mozilla!
http://quality.mozilla.org/
http://wiki.mozilla.org/Mozilla_QA_Community
irc://irc.mozilla.org/qa - /nick mw22
Benjamin Smedberg
Jul 23, 2012, 2:26:42 PM7/23/12
to rob...@ocallahan.org, mozilla.dev.planning group
On 7/22/2012 10:16 PM, Robert O'Callahan wrote:
> The Flash 11.3 update has caused some very serious problems in Firefox.
> Repainting issues (bug 762948), issues resizing the browser window (bug
> 765788), and various crashes. I've heard reports of a spike in related
> activity on SUMO and reports of a lot of users leaving Firefox over these
> issues. So far these seem to be Adobe's bugs, probably related to their new
> sandbox since that is Firefox only (maybe Opera too?) and at least some of
> these bugs go away if the user disables the sandbox. We've looked into
> workarounds for some of them but it's not easy. Adobe is responding but not
> very quickly.
>
> Is anyone focused on this problem?
Yes, I am coordinating a team of people who have been working with Adobe
> The Flash 11.3 update has caused some very serious problems in Firefox.
> Repainting issues (bug 762948), issues resizing the browser window (bug
> 765788), and various crashes. I've heard reports of a spike in related
> activity on SUMO and reports of a lot of users leaving Firefox over these
> issues. So far these seem to be Adobe's bugs, probably related to their new
> sandbox since that is Firefox only (maybe Opera too?) and at least some of
> these bugs go away if the user disables the sandbox. We've looked into
> workarounds for some of them but it's not easy. Adobe is responding but not
> very quickly.
>
> Is anyone focused on this problem?
on these issues (or at least the most pressing ones). However, if we can
easily produce workarounds for bad Flash behavior in any of our bugs, we
should do so.
Basic crash/hang statistics are available from
https://crash-analysis.mozilla.com/bsmedberg/flash-summary.html
--BDS
PhillipJones
Jul 23, 2012, 8:26:42 PM7/23/12
to
Web site that use purely Flash (www.hallmark.com comes to mind) that
buttons meant to do something just sit and do nothing. the same sites,
in Chrome (I hate worth a passion but keep for testing purposes), iCab,
OperaNext, OmniWeb, Safari work perfectly. Only FF, SM, Camino, and
Aurora do not work.
--
Phillip M. Jones, C.E.T. "If it's Fixed, Don't Break it"
http://www.phillipmjones.net mailto:pjo...@kimbanet.com
Chris Ilias
Jul 24, 2012, 3:10:52 PM7/24/12
to
On 12-07-22 11:15 PM, akeybl wrote:
> Robert,
>
> The release management, stability, SUMO, metrics, and QA teams at Mozilla along with engineering (bsmedberg, etc) have all been actively engaged on issues related to, or suspected to be related to, Adobe Flash 11.3 since its release. Comments in the stability bugs you reference should reflect that. Similarly, Adobe has multiple teams engaged on 11.3 issues, and they are analyzing all actionable data we provide, working to fix as many of the top issues in new versions of Flash as possible.
>
> We weigh our options on a near daily basis, and update our support documentation with the best options for our affected users (including disabling protected mode). As you suggest, drastic solutions are drastic and should be avoided until all other better options have been exhausted.
>
> Rest assured that Flash 11.3 issues are getting the proper attention at Mozilla and will continue to until their resolution.
If/When there's a post-mortem about the this, could you please announce
> Robert,
>
> The release management, stability, SUMO, metrics, and QA teams at Mozilla along with engineering (bsmedberg, etc) have all been actively engaged on issues related to, or suspected to be related to, Adobe Flash 11.3 since its release. Comments in the stability bugs you reference should reflect that. Similarly, Adobe has multiple teams engaged on 11.3 issues, and they are analyzing all actionable data we provide, working to fix as many of the top issues in new versions of Flash as possible.
>
> We weigh our options on a near daily basis, and update our support documentation with the best options for our affected users (including disabling protected mode). As you suggest, drastic solutions are drastic and should be avoided until all other better options have been exhausted.
>
> Rest assured that Flash 11.3 issues are getting the proper attention at Mozilla and will continue to until their resolution.
it in this newsgroup?
Tanvi Vyas
Jul 24, 2012, 4:53:04 PM7/24/12
to dev-pl...@lists.mozilla.org
I agree with Dao. Instead of continuing to lose users (or potentially
lose users since we don't have numbers right now), can we disable
sandboxing until Adobe comes up with a fix? In the meantime, the users
are no worse off than they were a few weeks ago before 11.3 came out.
In a week or two when the update is available, users will upgrade to a
version with Protected Mode enabled that does not crash Firefox.
~Tanvi
lose users since we don't have numbers right now), can we disable
sandboxing until Adobe comes up with a fix? In the meantime, the users
are no worse off than they were a few weeks ago before 11.3 came out.
In a week or two when the update is available, users will upgrade to a
version with Protected Mode enabled that does not crash Firefox.
~Tanvi
Alex Keybl
Jul 24, 2012, 6:31:54 PM7/24/12
to Chris Ilias, dev-pl...@lists.mozilla.org
> If/When there's a post-mortem about the this, could you please announce it in this newsgroup?
We'll make sure to announce any meetings we have on this subject.
-Alex
Martijn
Jul 30, 2012, 11:28:53 AM7/30/12
to Benjamin Smedberg, mozilla.dev.planning group, rob...@ocallahan.org
On Mon, Jul 23, 2012 at 8:26 PM, Benjamin Smedberg
<benj...@smedbergs.us> wrote:
> On 7/22/2012 10:16 PM, Robert O'Callahan wrote:
>>
>> The Flash 11.3 update has caused some very serious problems in Firefox.
>> Repainting issues (bug 762948), issues resizing the browser window (bug
>> 765788), and various crashes. I've heard reports of a spike in related
>> activity on SUMO and reports of a lot of users leaving Firefox over these
>> issues. So far these seem to be Adobe's bugs, probably related to their
>> new
>> sandbox since that is Firefox only (maybe Opera too?) and at least some of
>> these bugs go away if the user disables the sandbox. We've looked into
>> workarounds for some of them but it's not easy. Adobe is responding but
>> not
>> very quickly.
>>
>> Is anyone focused on this problem?
>
> Yes, I am coordinating a team of people who have been working with Adobe on
> these issues (or at least the most pressing ones). However, if we can easily
> produce workarounds for bad Flash behavior in any of our bugs, we should do
> so.
I can easily reproduce the Flash crash that is apparently related to
<benj...@smedbergs.us> wrote:
> On 7/22/2012 10:16 PM, Robert O'Callahan wrote:
>>
>> The Flash 11.3 update has caused some very serious problems in Firefox.
>> Repainting issues (bug 762948), issues resizing the browser window (bug
>> 765788), and various crashes. I've heard reports of a spike in related
>> activity on SUMO and reports of a lot of users leaving Firefox over these
>> issues. So far these seem to be Adobe's bugs, probably related to their
>> new
>> sandbox since that is Firefox only (maybe Opera too?) and at least some of
>> these bugs go away if the user disables the sandbox. We've looked into
>> workarounds for some of them but it's not easy. Adobe is responding but
>> not
>> very quickly.
>>
>> Is anyone focused on this problem?
>
> Yes, I am coordinating a team of people who have been working with Adobe on
> these issues (or at least the most pressing ones). However, if we can easily
> produce workarounds for bad Flash behavior in any of our bugs, we should do
> so.
https://bugzilla.mozilla.org/show_bug.cgi?id=773419
I get this crash a lot when watching a Flash video and then trying to
go in full screen mode. (Youtube comes to mind)
This is with the latest experimental Flash version: 11.3.300.268
Currently, I can't use Firefox to watch Flash videos in full screen mode.
Regards,
Martijn
> Basic crash/hang statistics are available from
> https://crash-analysis.mozilla.com/bsmedberg/flash-summary.html
>
> --BDS
>
>
> _______________________________________________
> dev-planning mailing list
> dev-pl...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-planning
> dev-planning mailing list
> dev-pl...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-planning
Robert Kaiser
Jul 31, 2012, 12:02:05 PM7/31/12
to
Martijn schrieb:
nowadays. :)
> This is with the latest experimental Flash version: 11.3.300.268
This is not experimental, it's their latest release. Their experimental
versions (available from Adobe Labs) are 11.4.* right now.
Robert Kaiser
> I can easily reproduce the Flash crash that is apparently related to
> https://bugzilla.mozilla.org/show_bug.cgi?id=773419
As you see there, Adobe people are actively engaged with Bugzilla
> https://bugzilla.mozilla.org/show_bug.cgi?id=773419
nowadays. :)
> This is with the latest experimental Flash version: 11.3.300.268
versions (available from Adobe Labs) are 11.4.* right now.
Robert Kaiser
Martijn
Aug 6, 2012, 3:51:39 PM8/6/12
to Robert Kaiser, dev-pl...@lists.mozilla.org
On Tue, Jul 31, 2012 at 6:02 PM, Robert Kaiser <ka...@kairo.at> wrote:
> Martijn schrieb:
>
>> I can easily reproduce the Flash crash that is apparently related to
>> https://bugzilla.mozilla.org/show_bug.cgi?id=773419
>
>
> As you see there, Adobe people are actively engaged with Bugzilla nowadays.
> :)
That's cool. Let's hope they will fix this bug soon, because it is
> Martijn schrieb:
>
>> I can easily reproduce the Flash crash that is apparently related to
>> https://bugzilla.mozilla.org/show_bug.cgi?id=773419
>
>
> As you see there, Adobe people are actively engaged with Bugzilla nowadays.
> :)
incredibly easy to hit this crash.
(and hopefully that would fix the general Flash crash spike)
Regards,
Martijn
>
>> This is with the latest experimental Flash version: 11.3.300.268
>
>
> This is not experimental, it's their latest release. Their experimental
> versions (available from Adobe Labs) are 11.4.* right now.
>
> Robert Kaiser
>
>
Robert Kaiser
Aug 6, 2012, 4:24:50 PM8/6/12
to
Martijn schrieb:
compared to other crash and hang signatures we're seeing with Flash, so
while it might help reduce the spike slightly, this by far won't fix it.
Robert Kaiser
> On Tue, Jul 31, 2012 at 6:02 PM, Robert Kaiser <ka...@kairo.at> wrote:
>> Martijn schrieb:
>>
>>> I can easily reproduce the Flash crash that is apparently related to
>>> https://bugzilla.mozilla.org/show_bug.cgi?id=773419
>>
>>
>> As you see there, Adobe people are actively engaged with Bugzilla nowadays.
>> :)
>
> That's cool. Let's hope they will fix this bug soon, because it is
> incredibly easy to hit this crash.
> (and hopefully that would fix the general Flash crash spike)
Sorry to disappoint you on the latter, but this is quite low in volume
>> Martijn schrieb:
>>
>>> I can easily reproduce the Flash crash that is apparently related to
>>> https://bugzilla.mozilla.org/show_bug.cgi?id=773419
>>
>>
>> As you see there, Adobe people are actively engaged with Bugzilla nowadays.
>> :)
>
> That's cool. Let's hope they will fix this bug soon, because it is
> incredibly easy to hit this crash.
> (and hopefully that would fix the general Flash crash spike)
compared to other crash and hang signatures we're seeing with Flash, so
while it might help reduce the spike slightly, this by far won't fix it.
Robert Kaiser
0 new messages